Examining subpar park experiences through mixed-methods social media analysis

Abstract

ABSTRACT Rationale/Purpose: Urban parks and performance venues provide multiple benefits for residents and tourists. As park managers seek to continually improve the visitor experience, social media is a useful data source to capture visitor sentiment, particularly surrounding subpar experiences. Design/Methodology/Approach: In this case study, we used a mixed methods approach to understand the online image of Wolf Trap National Park for the Performing Arts, Virginia, USA. Ten years of Google Review data (n = 3289 reviews) were scraped and analyzed using Natural Language Processing techniques, including sentiment analysis and topic modeling. Deductive qualitative analysis further explored topics for additional trustworthiness and insight. Findings: Reviews were generally positive (81.33%). Bigram and sentiment analysis illustrated parking and transit-related concerns were the most salient within visitors’ subpar reviews. This was further supported through topic modeling and qualitative analyses. Practical Implications: Results inform managers of themes within online reviews and parameterize the overall experience and online image of the park. Findings are transferable to other park and performance venues seeking to meet diverse visitors’ needs. Research Contribution: This study evidences integration of different analytical techniques to understand visitor experiences in sport and leisure contexts, with a particular focus on subpar experiences.